U.S. patent application number 11/660712 was filed with the patent office on 2007-11-08 for systems and methods for multiplexing qkd channels.
This patent application is currently assigned to MAGIO TECHNOLIGIES, INC.. Invention is credited to J. Howell Mitchell, Harry Vig.
Application Number | 20070258592 11/660712 |
Document ID | / |
Family ID | 36036812 |
Filed Date | 2007-11-08 |
United States Patent
Application |
20070258592 |
Kind Code |
A1 |
Mitchell; J. Howell ; et
al. |
November 8, 2007 |
Systems and Methods for Multiplexing Qkd Channels
Abstract
Systems and methods for multiplexing two or more channels of a
quantum key distribution (QKD) system are disclosed. A method
includes putting the optical public channel signal (SP1) in
return-to-zero (RZ) format in a transmitter (T) in one QKD station
(Alice) and amplifying this signal (thereby forming SP1*) just
prior to this signal being detected with a detector (30) in a
receiver (R) at the other QKD station (Bob). The method further
includes precisely gating the detector via a gating element (40)
and a coincident signal (PN1') with pulses that coincide with the
expected arrival times of the pulses in the detected (electrical)
public channel signal (SP2). This allows for the public channel
signal to have much less power, making it more amenable for
multiplexing with the other QKD signals.
Inventors: |
Mitchell; J. Howell;
(Amherst, NH) ; Vig; Harry; (North Billerica,
MA) |
Correspondence
Address: |
OPTICUS IP LAW, PLLC
7791 ALISTER MACKENZIE DRIVE
SARASOTA
FL
34240
US
|
Assignee: |
MAGIO TECHNOLIGIES, INC.
|
Family ID: |
36036812 |
Appl. No.: |
11/660712 |
Filed: |
August 23, 2005 |
PCT Filed: |
August 23, 2005 |
PCT NO: |
PCT/US05/29893 |
371 Date: |
February 21, 2007 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
60607540 |
Sep 7, 2004 |
|
|
|
Current U.S.
Class: |
380/259 |
Current CPC
Class: |
H04L 9/0852 20130101;
H04B 10/70 20130101 |
Class at
Publication: |
380/259 |
International
Class: |
H04L 9/00 20060101
H04L009/00; H04K 3/00 20060101 H04K003/00 |
Claims
1. A method of sending a public channel signal (SP1) and a quantum
channel signal (SQ) over an optical fiber connecting first and
second quantum key distribution (QKD) stations, comprising: a) at
the first QKD station: generating signal SQ signal and generating
SP1 in a return-to-zero (RZ) format; transmitting signals SP1 and
SQ over the optical fiber to the second QKD station; b) at the
second QKD station: demultiplexing signals SP1 and SQ; optically
amplifying signal SP1 to form an amplified signal SP1*; detecting
and processing amplified signals SP1* to form a frequency-locked
gating signal; and using the frequency-locked gating signal to gate
further detection of signals SP1* so as to reduce public channel
signal detection noise.
2. The method of claim 1, wherein the public channel signal SP1 is
formed from a 10 MHz Ethernet signal.
3. The method of claim 1, further including: detecting signal SP1*
to create a public channel electrical signal SP2; forming from
signal SP2 a signal PN1' that comprises electrical pulses that are
frequency locked and that are coincident with signal SP2; and using
signal PN1' to gate the detecting of signals SP1*.
4. The method of claim 3, wherein forming signal PN1' includes:
sending signal SP2 through a narrow bandpass filter to create a
sine-wave signal S3 that is frequency locked to signal SP2; passing
signal S3 through a comparator to create a square-wave signal S4;
forming signal PN1 by passing signal S4 through a multi-vibrator;
and passing signal PN1 through a delay controlled by a controller
adapted to provide signal PN1 with a delay in forming coincident
signal PN1'.
5. The method of claim 4, wherein the public channel signal is a 10
MHz Ethernet signal and the narrow bandpass filter has a bandwidth
of 10 MHz.
6. The method of claim 3, wherein forming signal PN1' includes:
passing signal PN1 through a variable delay line; and controlling
the variable delay line via a programmable controller operably
coupled thereto and adapted to impose a delay on signal PN1 to form
coincident signal PN1'.
7. The method of claim 6, including providing the controller with a
cross-correlation signal SC that provides information about an
amount of delay to be provided by the variable delay line.
8. The method of claim 7, including providing the controller with a
field-programmable gate array (FPGA) programmed with logic adapted
to find a maximum SCmax in signal SC.
9. A method of sending a public channel optical signal SP1 from a
transmitter in a first QKD station to a receiver in second QKD
station, comprising: providing signal SP1 in a return-to-zero (RZ)
format in the transmitter; sending signal SP1 from the transmitter
to the receiver; at the receiver, optically amplifying signal SP1
to form an optically amplified signal SP1* and detecting signal
SP1* to create a public channel electrical signal SP2; and
processing the public channel electrical signal SP2 to form a
frequency-locked coincident gating signal PN1' that gates the
detection of amplified signals SP1* so as to reduce public channel
detection noise.
10. The method of claim 9, further including: forming from signal
SP2 a signal PN1 comprising a train of narrow square wave
electrical pulses that are frequency locked with signal SP2;
cross-correlating signal SP2 with signal PN1 to create a
cross-correlation signal SC; finding a maximum of signal SC as
SCmax; and using SCmax to ensure that signal PN1 coincides with
signal SP2 to form signal PN1'.
11. The method of claim 10, wherein forming coincident pulses
(signal) PN1' includes: sending signal SP2 through a
narrow-bandpass filter to create a sine-wave signal S3 that is
frequency locked to signal SP2; passing signal S3 through a
comparator to create a square-wave signal S4; forming signal PN1 by
passing signal S4 through a multi-vibrator; and passing signal PN1
through a delay controlled by a controller adapted to provide
signal PN1 with a delay that makes signal PN1' coincident with
signal SP2.
12. A method of reducing detection noise in a public channel signal
(SP1) of a first wavelength multiplexed with quantum channel signal
(SQ) of a second wavelength as part of a quantum key distribution
(QKD) system having first and second QKD stations, comprising: at a
transmitter in the first QKD station, providing signal SP1 in a
return-to-zero (RZ) format, multiplexing public channel signal SP1
with quantum signal SQ, and sending the multiplexed signals to a
receiver in the second QKD station; at the receiver, optically
amplifying the demultiplexed signal SP1 to form an optically
amplified signal SP1*; detecting signal SP1* to create a public
channel electrical signal SP2; forming from signal SP2 a signal
PN1' that comprises electrical pulses that are frequency locked and
that are coincident with signal SP2; and using signal PN1' to gate
the detecting of signals SP1* to reduce detection noise associated
with public channel signal SP1.
13. A method of transmitting a public channel signal SP1 from a
first QKD station to a second QKD station over an optical fiber,
including: putting signal SP1 in return-to-zero (RZ) format and
transmitting the RZ signal SP1 to the second QKD station;
amplifying the RZ signal SP1 at the second QKD station to form an
amplified public channel signal SP1*; detecting signal SP1* with a
detector; and processing signal SP1* to establish a gating signal
PN1' that gates the detector to coincide with the expected arrival
times of signals SP1* in order to reduce public channel signal
detection noise.
14. The method of claim 13, including forming gating signal PN1' as
a train of pulses that are frequency-locked with an electrical
public channel signal SP2 formed by detecting signals SP1*, and
applying a selective delay to the train of pulses so that the
gating signal PN1' coincides in time with the electrical public
channel signal SP2.
15. An apparatus for sending and receiving a public channel signal
between first and second quantum key distribution stations,
comprising: A return-to-zero (RZ) encoder adapted to receive a
non-RZ signal SE and form therefrom an RZ format signal S6; a light
source system operably coupled to the RZ encoder and that generates
an RZ optical public channel signal (SP1) corresponding to RZ
format signal S6; an optical amplifier arranged to receive the RZ
public channel signal SP1 and form therefrom an amplified public
channel signal SP1*; a detector arranged to detect the amplified
public channel signal SP1* and generate therefrom a public channel
electrical signal SP2; and signal processing electronics adapted to
receive the public channel electrical signal SP2 and form therefrom
a frequency locked gating signal PN1' that coincides in time with
signal SP2 and that is used to gate the detector in order to reduce
public channel detection noise.
16. The apparatus of claim 15, further including a gating element
operably coupled to the detector and that is adapted to receive
gating signal PN1' so as to gate public channel electrical signals
SP2 generated by the detector.
17. A public channel receiver for a quantum key distribution (QKD)
station, comprising: an optical amplifier adapted to receive and
amplify an return-to-zero (RZ) optical public channel signal; a
detector adapted to generate a detector electrical signal at a
detector output in response to detecting the amplified public
channel signal; a gating element operably coupled to the detector
output and adapted to gate the detector output based on a gating
signal; a narrow bandpass filter coupled to an output of the gating
element and adapted to form from the detector electrical signal a
sine-wave signal that is frequency-locked to the detector
electrical signal; a comparator coupled to an output of the
narrow-band filter at a first comparator input and to a threshold
signal at a second comparator input, the comparator adapted to form
at a comparator output a square-wave signal from the inputted
sine-wave and threshold signals; a multi-vibrator coupled to the
comparator output and adapted to form a train of pulses from the
square-wave signal at a multi-vibrator output; a variable delay
line coupled to the multi-vibrator output and adapted to provide
the train of pulses with a select delay; a multiplier adapted to
receive the selectively delayed train of pulses and the detector
electrical signal and form therefrom a cross-correlation signal;
and wherein the cross-correlation signal is used to define the
select delay that causes the train of pulses to coincide with the
detector electrical pulses, and wherein the coincident train of
pulses is provided to the gating element to gate the detector.
18. An receiver apparatus for a first quantum key distribution
(QKD) station that allows for the transmission of a return-to-zero
(RZ) optical public channel signal and an optical quantum channel
signal over an optical fiber connecting the first QKD station to a
second QKD station, comprising: an optical amplifier adapted to
optically amplify the optical public channel signal provided in RZ
format; a detector adapted to generate from the amplified optical
public channel signal a detector electrical signal; a narrow-band
filter adapted to form from the detector electrical signal a
sine-wave signal frequency-locked to the detector electrical
signal; a comparator adapted to form at a second output a
square-wave signal from the sine-wave signal; a multi-vibrator
adapted to form at train of pulses from the square-wave signal at a
third output; a variable delay line coupled to the multi-vibrator
output and adapted to provide a select delay to the train of pulses
at a fourth output; a multiplier adapted to receive the selectively
delayed train of pulses and the detector electrical signal and
generate therefrom a cross-correlation signal; wherein the
cross-correlation signal is used to determine the selective delay
that causes the train of pulses to coincide with the detector
electrical pulse; and wherein the selectively delayed train of
pulses is used to control a gate arranged between the detector and
the narrow-band filter.
19. An apparatus for transmitting a public channel signal SP1 and a
quantum signal SQ over a single optical fiber, comprising: means
for generating a return-to-zero (RZ) signal SP1; means for
multiplexing signals SP1 and SQ onto the optical fiber; means for
demultiplexing signals SP1 and SQ; means for amplifying signal SP1;
means for detecting amplified optical signal SP1 and generating
therefrom an electrical signal (SP2); signal processing means for
receiving signal SP2 and forming therefrom a gating signal that
coincides with signal SP2; and gating means for gating the detector
using the gating signal.
Description
CLAIM OF PRIORITY
[0001] This application claims the benefit of priority under 35
U.S.C. .sctn. 119(e) of U.S. Provisional Patent Application Ser.
No. 60/607,540, filed on Sep. 7, 2004.
TECHNICAL FIELD OF THE INVENTION
[0002] The present invention relates to and has industrial utility
with respect to quantum cryptography, and in particular relates to
and has industrial utility with respect to multiplexing different
channels of a QKD system onto a single optical fiber.
BACKGROUND OF THE INVENTION
[0003] Quantum key distribution involves establishing a key between
a sender ("Alice") and a receiver ("Bob") by using weak (e.g., 0.1
photon on average) pulsed optical signals transmitted over a
"quantum channel." The security of the key distribution is based on
the quantum mechanical principle that any measurement of a quantum
system in unknown state will modify its state. As a consequence, an
eavesdropper ("Eve") that attempts to intercept or otherwise
measure the quantum signal will introduce errors into the
transmitted signals and reveal her presence.
[0004] The general principles of quantum cryptography were first
set forth by Bennett and Brassard in their article "Quantum
Cryptography: Public key distribution and coin tossing,"
Proceedings of the International Conference on Computers, Systems
and Signal Processing, Bangalore, India, 1984, pp. 175-179 (IEEE,
New York, 1984). Specific QKD systems are described in U.S. Pat.
No. 5,307,410 to Bennett, and in the article by C. H. Bennett
entitled "Quantum Cryptography Using Any Two Non-Orthogonal
States", Phys. Rev. Lett. 68 3121 (1992). The general process for
performing QKD is described in the book by Bouwmeester et al., "The
Physics of Quantum Information," Springer-Verlag 2001, in Section
2.3, pages 27-33.
[0005] The performance of a QKD system can be degraded by noise in
the form of photons generated by three different mechanisms. The
first is forward Raman scattering, in which frequency-shifted
photons are generated and co-propagate with the quantum signal
photons. Raman scattering in an optical fiber limits the power that
can be put into a single fiber because of a transfer of energy from
a high power signal to the single-photon wavelength.
[0006] The second mechanism is Raman backscattering, in which
frequency-shifted photons are generated and propagate in the
opposite direction to the quantum signal photons.
[0007] The third mechanism is Rayleigh scattering, in which photons
are elastically scattered back in the opposite direction of the
quantum signal photons.
[0008] The scattering of light in an optical fiber--and particular
forward Raman scattering--is problematic in multiplexing the
different channels of a QKD system because of the noise it creates
in the detection process.
[0009] Two simple solutions have been proposed to overcome the
effects of light scattering in combining different channels onto a
single optical fiber. The first solution is to use one fiber for
the public discussion channel, possibly the sync channel as well,
and a second fiber for the quantum channel. The second solution is
to limit the fiber length so that the input power can be reduced,
and so the scattering power transfer ratio is lower with the
shorter distance. Both of these solutions, while simple, are also
unappealing because they are not particularly robust and are
ill-suited for a commercially viable QKD system.
[0010] The prior art relating to multiplexing the different
channels associated with QKD includes U.S. Pat. No. 6,438,234 ("the
'234 patent"). In the '234 patent, the sync signal is
time-multiplexed with the quantum channel. The prior art also
includes U.S. Pat. No. 5,675,648 ("the '648 patent). The '648
patent proposes the idea of having a "common transmission medium"
(i.e., an optical fiber) for the quantum channel and the public
channel, where the public channel also carries a calibration
signal.
[0011] However, the prior art does not address the daunting problem
of combining the relatively strong public and sync channels with
the very weak quantum channel. In particular, the '648 patent does
not address how the public channel can be multiplexed with the
quantum channel in the "common transmission medium" in a way that
will not interfere with detecting the single-photons associated
with the quantum channel.
[0012] Also, in the '234 patent, a sample-and-hold type of phase
lock loop needs to be implemented to hold the sync timing while
working on single photons. However, the difficulties of
multiplexing sync and quantum channel are less challenging than the
task of multiplexing the public (data) channel and the quantum
channel. The '234 patent does not address the issue of transmitting
the public channel and the quantum channel over the same optical
fiber.
[0013] The publication "Eighty kilometer transmission experiment
using an InGaAs/InP SPAD-based quantum cryptography receiver
operating at 1.55 um" by P. A. Hiskett, G. Bonfrate, G. S. Buller,
and P. D. Townsend, published in the Journal of Modern Optics,
2001, vol. 48, no. 13, pp. 1957-1966, suggests an approach to
combining the sync and quantum channels. The light from the
transmitter laser is split into a quantum signal and a sync signal.
The sync signal is sent over a separate fiber and upon entering the
receiver is amplified by an erbium doped fiber amplifier (EDFA).
After the amplified light signal is converted into electrical
signal, the electrical signal is used to gate the receiver's
detector.
[0014] It would be desirable to wavelength-multiplex 10 MHz
Ethernet public discussion traffic (i.e., the public channel) onto
the same fiber as the sync and quantum channels. However, the
optical power of the Ethernet public channel signal must be
significantly reduced to prevent scattering and other such
interference that reduces the ability to detect the channels.
Unfortunately, reducing the public channel power results in an
unacceptably low signal-to-noise ratio for the public channel for
any QKD system with a satisfactory distance or span. While the use
of an optical fiber amplifier (e.g., an erbium-doped fiber
amplifier or EDFA) can increase the amplitude of the optical signal
and remove the need for a narrow band optical filter, its output
will still have a very low signal to noise ratio.
DESCRIPTION OF THE INVENTION
[0015] The present invention includes systems and methods for
multiplexing two or more channels of a quantum key distribution
(QKD) system. The systems and method result in reduced detection
noise for the public channel, thereby allowing weaker public
channel signal to be used. Use of a weaker public channel signal
enables multiplexing the public channel with the quantum channel
and/or the sync channel on the same optical fiber for a
commercially viable QKD system.
[0016] An aspect of the invention is a method that includes putting
the optical public channel signal in return-to-zero (RZ) format and
amplifying this signal just prior to it being detected. The method
further includes precisely gating the detector to coincide with the
expected arrival times of the pulses in the detected (electrical)
public channel signal to reduce the detection noise. The method
also includes forming a first signal comprising a train of pulses
that is frequency-locked with the electrical public channel signal,
and then applying a selective delay to the first signal so that the
first signal coincides (in time) with the electrical public channel
signal. The first signal is then used to gate the detector.
[0017] This method serves to drastically reduce the noise
associated with the detection of the public channel, which in turn
allows for a weaker public channel signal to be used. Use of a
weaker public channel signal is what enables multiplexing the
public channel with the quantum channel and/or the sync channel on
the same optical fiber.
[0018] The method is generally applicable to detecting a weakened
Ethernet signal that would otherwise be difficult to detect.
[0019] Another aspect of the invention is a method of combining a
public channel signal (SP1) of a first wavelength and a quantum
channel signal (SQ) of a second wavelength on an optical fiber
connecting first and second quantum key distribution (QKD) stations
(Alice and Bob). The method includes providing signal SP1 in a
return-to-zero (RZ) format and multiplexing and transmitting
signals SP1 and SQ from Alice to Bob. The method also includes
wavelength-demultiplexing signals SP1 and SQ at Bob and optically
amplifying signal SP1 to form an optically amplified signal SP1*.
The method also includes detecting signal SP1* to create a public
channel electrical signal SP2 and forming from this signal a signal
PN1' that comprises electrical pulses that are frequency-locked and
coincident (in time) with signal SP2. Finally, the method includes
using signal PN1' to gate the detecting of signals SP1*.
BRIEF DESCRIPTION OF THE DRAWINGS
[0020] FIG. 1 is a schematic diagram of a transmitter-receiver
(T-R) system for use with QKD stations of a QKD system in order to
effectively transmit the public channel along with the quantum
channel and/or the sync channel on the same optical fiber;
[0021] FIG. 2 is a schematic diagram of a QKD system with two QKD
stations Alice and Bob showing how the T-R system is used in a QKD
system, wherein Alice has a transmitter T1 and a receiver R2, and
Bob has a transmitter T2 and a receiver R1, so that two-way
communication over the public channel is enabled by T1-R1 and
T2-R2.
[0022] The various elements depicted in the drawings are merely
representational and are not necessarily drawn to scale. Certain
sections thereof may be exaggerated, while others may be minimized.
The drawings are intended to illustrate various embodiments of the
invention that can be understood and appropriately carried out by
those of ordinary skill in the art.
DETAILED DESCRIPTION OF THE BEST MODE OF THE INVENTION
[0023] The present invention relates to quantum cryptography, and
in particular relates to systems and methods that allow for select
channels of a QKD system, such as the public discussion channel,
the synchronization ("sync") channel and/or the quantum channel, to
be multiplexed on to a single optical fiber. In the discussion
herein, the quantum channel carries the quantum signal, which is a
single-photon optical pulse. The term "single photon" is meant to
encompass optical pulses having one photon or less on average.
[0024] The sync channel as discussed herein carries synchronization
data (signals), and optionally, calibration data (signals) that
allows for the coordinated operation between the two QKD stations,
which are typically denoted as Bob and Alice.
[0025] Also in the discussion below, the terms "signal" and "pulse"
are used interchangeably in a manner that will be apparent to those
skilled in the art. Also, the terms "public channel signal" and
"quantum signal" are each understood as including one or more
pulses, e.g., a train of pulses.
[0026] For commercial QKD systems, there is a strong desire to use
an existing optical fiber to carry two or more of the QKD channels
between QKD stations. The present invention enables carrying all
three of the above-mentioned channels on a relatively long optical
fiber (e.g., 50 km to 100 km) normally associated with a
commercially viable QKD system.
[0027] Note that in a typical QKD system, the two QKD station are
referred to as "Alice" and "Bob," and transmission occurs over the
QKD channel in one direction, i.e., from Alice to Bob. However, in
connection with an Ethernet public discussion channel, Alice and
Bob are identical peers. That is, in order to support the
Ethernet-related protocols (e.g. TCP/IP) over the public channel,
bi-directional communication is required. This, in turn, means that
Alice and Bob each contain a receiver R and a transmitter T, as
discussed below.
QKD System
[0028] FIG. 1 is a schematic diagram of an example embodiment of a
transmitter-receiver (T-R) system 2 according to the present
invention. The T-R system 2 includes a QKD station transmitter T
that is coupled to a QKD station receiver R by an optical fiber
link FL. FIG. 2 illustrates how the T-R system is incorporated into
a QKD system as two systems T1-R1 and T2-R2 to achieve
bi-directional public channel communication, as described in
greater detail below.
[0029] The transmitter T includes three light source systems L1, L2
and L3 operating at respective wavelengths .lamda.1, .lamda.2 and
.lamda.3. Light source systems L1, L2 and L3 are respectively
adapted to generate corresponding quantum signal SQ, sync signal SS
and public channel signal SP1. For example, light source system L3
is adapted to provide the public channel signal SP1 in a variety of
formats, including return-to-zero (RZ) format. Light source systems
L1, L2 and L3 are optically coupled with and wavelength-multiplexed
onto fiber link FL via a wavelength-division multiplexer 5.
[0030] In an example embodiment, the transmitter T includes an RZ
encoder 6 that accepts an industry-standard 10 MHz Ethernet
Manchester-encoded signal SE from an Ethernet port EP1. RZ encoder
6 converts signal SE to narrow, low-duty-cycle pulses S6. Signals
S6 are then used to drive light source system L3 in order to
generate relative low-power optical public channel signals SP1 that
have a 10-MHz-Ethernet RZ format.
[0031] With continuing reference to FIG. 1, receiver R includes a
wavelength-division demultiplexer 8 optically coupled to optical
fiber link FL. Demultiplexer 8 is adapted to separate optical
signals SQ, SS and SP1 with wavelengths .lamda.1, .lamda.2 and
.lamda.3, into separate optical paths, e.g., separate optical fiber
sections. The two optical paths associated with quantum signal SQ
at wavelength .lamda.1 and sync signal SS at wavelength .lamda.2
are indicated by 9. The third optical path associated with the
public channel signal SP1 and wavelength .lamda.3 is indicated by
10.
[0032] Note that in FIG. 1, the details of quantum channel and the
sync channel apparatus are not shown in transmitter T and receiver
R and because they are not critical to the understanding of the
invention and are based on known art.
[0033] T-R system 2 of FIG. 1 further includes along optical path
10 (e.g., optical fiber section 10) downstream of
wavelength-division multiplexer 8 an optical amplifier 20, such as
an erbium-doped fiber amplifier (EDFA). Optical amplifier 20 is
adapted to optically amplify optical public channel signal SP1 to
form an amplified optical public channel signal SP1* just prior to
or soon after signal SP1 enters receiver R. Optical amplifier 20 is
shown within the receiver in FIG. 1 for the sake of
illustration.
[0034] Downstream of optical amplifier 20 is a detector 30 (e.g., a
PIN photodiode) operably coupled to the optical amplifier, and a
gating element ("gate") 40 (i.e., a fast on-off switch) downstream
and operably coupled to detector 30. The output of gate 40 is
coupled to a filter 50, which in the present example is a 10 MHz
narrow-bandpass filter.
[0035] The output of filter 50 is operably coupled to one input of
a high-speed comparator 60. The other input of comparator 60 is
provided with a threshold signal ST. The output of comparator 60 is
coupled to a multi-vibrator 65 (e.g., a one-shot or mono-stable
multi-vibrator). The output of multi-vibrator 65 is coupled to a
variable delay 70, which is controlled by a programmable controller
80 operatively coupled to the delay. In an example embodiment,
controller 80 includes a field-programmable gate array (FPGA). The
output of variable delay 70 is also coupled to gate 40 via line
72.
[0036] One of the outputs from variable delay 70 is connected to
one input port of a multiplier 90, while an input line 82 is
coupled to the other multiplier port. Line 82 carries the public
channel signals (pulses) SP2 that make it through gate 40, as
discussed below. The output of multiplier 90 is sent to the input
of a low-pass filter 100, whose output is connected to an input of
controller 80. Controller 80 then controls the variable delay 70,
which has an output to gate 40.
[0037] As mentioned above, in an example of the present invention,
public channel signal SP1 is a 10 MHz Ethernet Manchester-encoded
data stream re-coded into an RZ format with very narrow RZ pulses.
This allows the output of optical amplifier 20 to be gated (or
enabled) via variable delay 70 to the multiplier (90) only when the
RZ pulses might be present. The presence of a narrow pulse
represents a data bit of "1" and the lack of a narrow pulse
represents a "0". The narrow RZ pulses occur only on the Ethernet
10 MHz square wave edges.
[0038] The gating of the optical amplifier output significantly
reduces the noise in the public channel signal detection process.
However, such gating requires that the time slots where the narrow
RZ pulses occur be known. Fortunately, the frequency of the public
channel signals is known to within 100 PPM (Parts Per Million), as
is consistent with the IEEE 802.3 standard. This information is
used to produce the required detector gating signal in the manner
described below.
Method of Operation
[0039] An example embodiment of the present invention uses a
non-return-to-zero (NRZ) Manchester-encoded and industry-standard
10 MHz Ethernet signal and converts it to an RZ format using RZ
encoder 6. The resulting RZ public channel signal SP1 is then sent
over the public channel, as mentioned above. Public channel signal
SP1 is multiplexed with the quantum and sync channel signals SQ and
SC via multiplexer 5, and sent over to receiver R via optical fiber
link FL. The public channel signal S1 is then demultiplexed from
the quantum signal and sync signals (not shown) by demultiplexer 8
and is amplified by optical amplifier 20 to form amplified public
channel signal SP1*. The amplified signal SP1* is then detected by
detector 30, which converts this signal into a corresponding
electrical public signal SP2.
[0040] The electrical public signal SP2 passes through gate 40
(whose operation is discussed below) and travels to filter 50
(e.g., a 10 MHz bandpass filter). Filter 50 creates a (10 MHz)
sine-wave signal S3 that is frequency-locked to the incoming
Ethernet RZ data (i.e., electrical public signal SP2).
[0041] High-speed comparator 60 receives sine-wave signal S3 at the
"+" input and the threshold signal ST at the "-" input, and
converts signal S3 to a (10 MHz) square wave signal S4 at the
comparator output. The square-wave signal S4 then passes to
multi-vibrator 65, which converts the signal to a train of narrow
electrical signals (pulses) PN1. The pulse width of multi-vibrator
65 is preferably as great or slightly greater than the width of
signal SP2 that travels through gate 40.
[0042] Pulses PN1 enter delay 70, whose delay is selectively
controlled by programmable controller 80. It is the job of
controller 80 to impart a selective delay to pulses PN1 so they
fall directly on top of (i.e., are coincident in time with) the
incoming narrow Ethernet RZ signals SP2. For the sake of clarity,
the train of selectively delayed pulses created by delay 70 are
referred to as signal PN1'.
[0043] Signal PN1' from variable delay 70 is multiplied with the
incoming RZ Ethernet pulses (i.e., electrical public channel signal
SP2) from input line 82 at multiplier 90. Multiplier 90 creates a
cross-correlation function signal SC from the two multiplier input
signals. Signal SC is provided to controller 80 through a low-pass
filter 100. In an example embodiment, it is assumed that controller
80 makes changes to the delay values slowly, because quick changes
could result in closed-loop instability. The controller only needs
to initially acquire and then track the input pulse train (i.e.,
signal SP2), neither of which requires a quick response. The low
pass filter 100 removes all of the high-speed information which is
of no value and that could destabilize the system. Also, note that
statistically half of the RZ signal SP2 (e.g., Ethernet RZ pulses)
are missing (for logic "0's"); the low pass filter is need to
"smooth over" these gaps.
[0044] In an example embodiment, an analog-to-digital (AD)
converter 101 is arranged between low-pass filter 100 and
controller 80 to create digital signals SC from analog signals SC
in the case where controller 80 is a digital device.
[0045] Based on the information in signal SC, controller 80
controls variable delay 70 via a control signal S5 to form
coincident signal PN1'. Signal PN1' is sent over line 72 to control
the operation of gate 40. In other words, coincident signal PN1' is
used as a gating signal to control the operation of gate 40.
[0046] If the output signals (pulses) PN1' from variable delay 70
and the RZ Ethernet pulses SP2 are in phase, then the multiplier
output signal SC will be at a maximum. In an example embodiment,
the cross-correlation of multiplier 90 is averaged over a time
period greater than a 10 MHz clock period (100 nanoseconds).
[0047] When signal SC is maximized, the pulses in delay output
signal PN1' coincide with the Ethernet RZ pulses SP2. Controller 80
can therefore send these coincident pulses over line 72 to gate 40
to enable the gated detection of the optically amplified electrical
public channel signal SP2.
[0048] If, during the gating signal PN1' at line 72, a pulse is
found at the output (line 82) of gate 40, then the result is an
Ethernet logical "1". If, during the gating signal PN1' at line 72,
no pulse is found at the output of the gate, then the result is an
Ethernet logical "0". The train of Ethernet logical "1's" and "0's"
are then serially combined to produce a Manchester-encoded signal
SP2 that can be processed by standard, commercially available
Ethernet integrated circuits.
[0049] The conversion from the narrow RZ pulses to the wide
Manchester-encoded pulses required by the 10 MHz Ethernet standard
is performed by a decoder 110 coupled to the output of gate 40.
This describes the required receiver. Decoder 110, in turn, is
coupled to an Ethernet port EP2 or other like device.
Bi-Directional Public Channel Communication
[0050] FIG. 1 shows an example of a T-R system 2 by which
Manchester-coded public channel data flows from transmitter T to
receiver R as signal SP1. However, for bi-directional operation of
the public channel, another set of transmitters and receivers is
needed to carry data the other way.
[0051] Accordingly, FIG. 2 is a schematic diagram illustrating an
example implementation of a QKD system with QKD stations Alice and
Bob, each having a transmitter T and a receiver R as illustrated in
FIG. 1, thereby enabling Alice and Bob have bi-directional public
channel communication. Specifically, Alice has a transmitter T1 and
a receiver R2, and Bob has a transmitter T2 and a receiver R1, so
that two T-R systems--T1-R1 and T2-R2 are present.
[0052] Alice is coupled to Ethernet port EP1 while Bob is coupled
to Ethernet port EP2. In the example QKD system of FIG. 2, the
quantum signal SQ and the sync signal SC travel in one direction
from Alice to Bob, while the public channel signal SP1 travels
bi-directionally from Alice to Bob and from Bob to Alice.
[0053] With reference again to FIG. 1, in an example embodiment,
controller 80 in receiver R includes programmable logic (e.g., a
logic-programmed FPGA) adapted to determine the peak (maximum) of
the averaged cross-correlation function signal SC. Once the peak is
found, it keeps the delay matched to the incoming pulse train PN1
so that the peak is maintained over time, even in the face of
varying influences such as temperature fluctuations.
[0054] The coincident gating of the detection of the public channel
signal serves to drastically reduce the amount of noise in the
public channel detection process. This allows the optical power
level of the optical public channel signal SP1 to be reduced to the
point that it can coexist on the same optical fiber as the quantum
and/or the sync channels.
[0055] The present invention is described above in connection with
a 10 MHz Ethernet signal as an example embodiment of public channel
signal SP1. However, the present invention is applicable to most
any kind of data transmission at most any data rate. For instance,
Sonet, 100 MHz Ethernet, 1G Ethernet, etc. would all apply. Also
mentioned above is Manchester encoded data, but the present
invention is not so limited and would apply, for example, to 8B/10B
coding and most any other type of coding.
[0056] Further, the present invention is generally applicable to
QKD systems, (including one-way and two-way QKD systems) and
generally to telecommunication applications. In particular,
although the method is eminently suited for multiplexing weak
(single-photon) optical signals with relatively strong Ethernet
optical signals, it can be applied to cases involving Ethernet
signals only. The present invention can be applied to situations
wherein a standard Ethernet signal has to travel longer distances
than anticipated, resulting in having to detect a relatively weak
Ethernet signal. The present invention can thus be used to increase
the detectability of the weakened Ethernet signal without the need
for amplification.
* * * * *